After two hydraulic oil flushes, I still have some milky oil. I cooked up this hair-brained idea to use 8lb of desiccant in a five-gallon bucket and use a 650gph fish pond aeration pump to circulate the dry air into the headspace. I plan on drawing the air into the bucket at the bottom, from the hydraulic tank, allowing the desiccant to dry the air, pulling the dry air off the top of the bucket and back into the tank in a closed-loop system. I figure this will take a couple hundred hours and will require stoking the cylinders and track rotations to be effective. I know headspace flushes work with enough time and airflow. What are everyone's thoughts on submerging the dry return air hose into the hydraulic oil to introduce it at the bottom of the tank? I worry about entraining air in the oil but don't know enough about fluid dynamics to know if it's possible without any real pressure.
I've often wondered if a cream separator, the same type of thing dairy farmers use could be used to remove water from oil. Basically they are a centrifuge.
I'm not sure how fast you would have to spin it to remove the water. I could use water removing filters, but they are expensive and don't remove much water. 8 lbs of media should remove a half-gallon of water without recharging them and the whole setup is $75. I figure it's worth a shot and imagine it would work better by aerating the oil but don't want to create an air problem.
A cream separator is more complicated than it needs to be for separating water or dirt from oil on this small scale. A simpler centrifuge that requires manual cleaning will work for this quantity of oil. A centrifuge will not remove all of the water, I believe. Probably enough for most purposes. I'd suggest keeping the oil as warm as possible, and using an aquarium air stone to bubble the air through the oil. It's all about surface area and contact with the oil. Headspace will take exponentially longer to remove the water from the oil. If you're worried about oil carrying over to the dessicant, just run it through an air filter, even a dry oil or hydraulic filter will remove most of the oil mist, and the dessicant is cheap enough to replace.
Hydraulic oils have emulsifiers in the oil that work to hold the water in suspension. When I was in the navy we had two types of centrifugal purifiers for the oils in the reduction gears. They were a Sharpels (sp) and DeLaval. The Sharpels was a vane type while the DeLaval was a plate type. The ran at around 25,000 RPM and were turned by three phase motors. I have no idea of what they cost, skies are the limit for military stuff, but I'm completely sure that they aren't cost effective. If you are going to try dry out the oil by moisture absorption into the air, you are going to have to heat it to near boiling and then hope to vent it while you are running the machine. Let me know if what you are trying works. I have several clients looking for an answer to the same problem.
I will report my findings with samples taken at different intervals similar to the link. The pic was after the first flush and just before I flushed and refilled the second time. I thought about trying vacuum dehydration with an a/c vacuum pump but don't want to risk any damage to anything. https://www.hydraulicsupermarket.com/blog/all/an-easy-way-to-remove-water-from-hydraulic-oil/
There was someone on youtube who posted a short segment of a video showing a kidney loop that they had rented or hired to clean the oil in a skidder. So the service is out there if you know where to look.
I know they're plenty of more expensive ways to remove water from oil. I figure for $75 it's worth a shot and maybe I can help the others here. There was a post on HEF a few years back about the cost of a vacuum dehydration machine and the math broken down on water removal filters and it was cost prohibitive.
The Navy type DeLaval separators that John C describes are very similar to the DeLaval cream separator (100HP version, not the hand cranked version), and I'll bet they ARE cost effective for the Navy, they probably put all their oils and fuels through them, and obviously aren't stopping things to clean the bowl for every gallon of water. Filters get expensive if there's much water, but vacuum dehydration does not have to be complicated. The hotter the oil gets, the less vacuum it takes to boil off the water. The link about headspace dehydration was interesting, the outside temp of 30 and the machine running most of the day were critical to that working so well.
We hook up a transducer to pull a vacuum on the tank and run the machine to get the oil hot as possible and cycle it. Another thing I have done is to use water separator filters on a hydraulic filter cart. any way you go it’s not fun or easy to get water out.
As far as I saw in the engine room, they only did the main engine and reduction gear oils. They got cleaned every eight hours when the engines were running. Main engines run on steam and you get condensation in the oils.
Day one of the test has started. I'll upload progress pictures over the next few days. I'm trying this with the oil cold as it's not practical to leave the machine running 24/7. I did notice I had pure water in where the strainer spring sits so I probably have a clogged vent.
I think I'd flush it again. As far as warmth get some of those magnetic tank heaters and let it cook.
Get the water out first, you'll never get the oil out by draining and refilling, and flushing is much more work than this. A 4,500 watt water heater element, will run just fine on 120V at 1,250 watts...
Checked everything at 40 hours of run time and I'm seeing improvement. I expect this will take a long time but it is working.
I worked at a place that would put the oil in drums and heat it up to boil the water off. They bought a centrifuge that was made to clean oil i never saw it because it need some work done on it and was sent to another shop were it was in a destroyed in a fire.